Metastatic relapse of breast cancer usually occurs several years after initial diagnosis and surgical treatment of the primary tumor. It is now widely assumed that, once tumor cells have extravasated in the target organ, they remain dormant for extended periods as a consequence of their inability to exit from quiescence. The signals that enforce dormancy as well as those that eventually enable a limited number of dormant cells to exit from their state and outgrow into macroscopic metastases are largely mysterious. Progress in this critical area of research has been hampered by the lack of mouse models amenable to genetic analysis. We have designed a gain-of-function screen that uses a mouse model of breast cancer dormancy in the lung as a filter to isolate cDNAs, microRNAs, and sh-RNAs that promote reactivation in this organ. The first cDNA, which we have isolated, encodes for the secreted antagonist of TGF-? ligands Coco. Coco induces metastasis-initiating cells (MICs) to exit from dormancy and undergo reactivation by blocking the ability of BMP proteins secreted by the lung stroma to inhibit their self-renewal. These results suggest that disseminated breast cancer cells require the self-renewal capability typically associated with stem cells in order to outgrow at metastatic sites. In addition, they imply that these cells need to overcome organ-specific anti-metastatic signals (Gao et al., Cell 2012, in press). By using the same approach, we have identified additional genes that mediate metastatic reactivation, such as the atypical tetraspanin TM4SF1, which appears to function by coupling the collagen receptor DDR1 to STAT3, and the microRNAs miR-138 and miR-346. In addition, we have identified the Notch inhibitor Numb as an enforcer of the dormant state. In this application, we will test the hypothesis that tumor dormancy and reactivation are governed by signalling pathways similar to those involved in stem cell renewal. Our Specific Aims are: 1) To Examine the Biochemical Basis of the Pro-metastatic Activity of Coco and the Preclinical Efficacy of Monoclonal Antibodies Blocking Coco; 2) To Elucidate the Mechanism by which the Atypical Tetraspanin TM4SF1 Promotes Metastatic Reactivation; 3) To Study the Mechanism by which the Asymmetric Cell Division Determinant Numb Enforces Tumor Dormancy; and 4) To Examine the Mechanism by which miR-138 and miR-346 Promote Metastatic Reactivation. By identifying the core signaling pathways that regulate breast cancer dormancy and reactivation, these studies will provide a rational framework to understand the logic of these processes. In addition, these studies are likely to lead to the identification of novel targets for therapeutic interventio as well novel mechanism-based biomarkers of metastatic potential.